Wind energy has become an important part of the power system due to its versatility,low cost,and environmental ***,the instability and volatility of wind energy pose a huge challenge to the grid-connected operation of...
详细信息
ISBN:
(数字)9789887581581
ISBN:
(纸本)9798350366907
Wind energy has become an important part of the power system due to its versatility,low cost,and environmental ***,the instability and volatility of wind energy pose a huge challenge to the grid-connected operation of wind power,affecting the effective scheduling of the power *** this situation,accurate forecasting of wind power becomes crucial,and deep neural networks have been widely used in this field due to their excellent ability to process high-dimensional nonlinear spatiotemporal ***,excellent deep neural network architectures are often manually designed by users with extensive expertise,which is time-consuming and results in high *** order to automatically design suitable deep neural network architectures for different wind power forecasting tasks,this paper proposes a deep learning wind power forecasting method based on architecture comparison evolutionary neural architecture *** method first encodes neural network architectures and then uses evolutionary algorithms to search for the optimal *** them,to address the problem of slow fitness calculation in evolutionary algorithms,architectures comparison-based the method is used instead of fitness calculation for architecture search,greatly accelerating search *** effectiveness and superiority of the proposed method were verified through case studies on actual wind farm datasets.
暂无评论